Learn graph
当用户要系统学一个新领域、不知道从哪入手、或担心「学得不够系统」时使用。用「知识图谱学习法」和用户一起构建该领域的概念/用途/父子节点图谱（自己建图的过程本身就是学习），标出复用价值最高的节点和「从常识就能入门的点」，给出有效学习路径并回答「学到哪算够」。触发场景：系统学 X 领域、从哪开始学、学得不系统、想要 X 的全貌、规划学习路径、这个领域有多大。From its SKILL.md
npx -y skills add Li-Evan/Bloom --skill learn-graphAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
SKILL.md
2.7 KB, 821 tokens by cl100k_base, as published. Nobody here has run it
知识图谱学习法（learn-graph）
核心信条：自己一步步建图谱的过程，本身就是最有效的学习——不要直接套用别人给的图谱。 绝大部分知识，都有一个从常识就能入门的点。
何时用
用户要系统进入一个新领域，或焦虑"学得不够系统 / 不知何时算够"。
流程（关键：和用户一起建，不是直接灌一张完整图）
第一步：锁定目标领域 X 和目的
用户为什么学 X？（接 learn-occam 的"既定问题"）目的决定图谱画到多细。
第二步：构建图谱——只抓三件事
概念/名称 · 用途 · 上下文关系（父子节点）：
- 子节点 = X 依托 / 基于什么；父节点 = X 服务于什么目标。
- 以提问引导用户一起填（自己建图才学得到），别一次性灌完。先给骨架，留节点让他补。
第三步：标注两个关键
- 复用价值：哪些节点父节点多（像 Python）→ 优先学，回报最高。
- 入门点：哪个节点"从常识就能入门"→ 学习路径的起点。
第四步：输出学习路径 + 颗粒度
从入门点出发、沿父子关系排一条有效路径。颗粒度按需自由切换（领域图 → 细分学科图）。"学到哪算够"= 覆盖到能解决第一步那个目的的节点即可，不必学满。
第五步：交接
- 拿不准某节点是不是缺前置知识 → 这正是图谱的强项，已在图上标出。
- 找到入门点要动手 → 转
learn-prototype（在图上找"最垃圾原型"的起点）。
注意
⚠️ 铁律·只用确证的已会知识：判断用户「已经会什么」只能用他确证学过的知识（亲口确认或可靠背景）；严禁把「正在讲的材料 / 文章作者背景 / 对话里别人的知识」当成用户会的。拿不准 → 直接问「⚠️ 你学过 ___ 吗？」，绝不替他假设。
- 强调"自己建"：多用提问让用户参与，别炫一张完美的图。
- 同族 skill：
learn-occam(该不该学)learn-crossover(已会什么)learn-prototype(动手)learn-feynman(自查)。
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.
Gives 0 of the 12 instructions most learn study skills give in 821 tokens
Counted across 545 of the 593 authors here whose files we hold, read 2026-09-06
- Treat the current directory as a teaching workspacein 20 of 545, across 17 files
- Teach knowledge first then practice skillsin 19 of 545, across 16 files
- Design lessons which build long-term retentionin 15 of 545, across 12 files
- Save each lesson as a self-contained HTML filein 15 of 545, across 12 files
- Question the user on why they want to learn thisin 15 of 545, across 12 files
- Reuse components from the assets directoryin 14 of 545, across 11 files
- Never trust your parametric knowledgein 13 of 545, across 10 files
- Record user preferences in NOTES.mdin 11 of 545, across 8 files
- Ground all teaching in the MISSION.md documentin 11 of 545, across 8 files
- Save each lesson to the lessons directoryin 8 of 545
- Question the user if the mission is unclearin 7 of 545
- Gather primary sources onlyin 7 of 545, across 4 files
Said here and by no other author read
- Lock target domain and purpose first
- Build graph focusing on concept purpose and context
- Mark reuse value and common sense entry point
- Output learning path and granularity
- Guide user to build graph together
- Ask when unsure about user knowledge
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.